Pith. sign in

REVIEW 3 major objections 6 minor 2 cited by

Quantized symbolic time series approximation

T0 review · 3 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Storing symbolic time-series centers as low-bit integers halves storage and keeps reconstruction accuracy, with a proved bound on the added error.

desk verdict A straightforward, useful storage optimization for ABBA with a sound quantization error bound; the 'retains accuracy' claim is overstated and the LLM 'SOTA' claim needs a real baseline. read the letter →

arxiv 2411.15209 v2 pith:IPNYRNRP submitted 2024-11-20 cs.LG eess.SPstat.ML

Erin Carson , Xinye Chen , Cheng Kang This is my paper · ORCID
classification cs.LGeess.SPstat.ML
keywords quantizedsymbolicapproximationABBAtimeseriesrepresentationquantizationerrorboundstorageefficiencyregressionlargelanguagemodels
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper argues that the symbolic centers of ABBA—the length and increment pairs defining each pattern—can be stored as low-bit-width integers with almost no loss in reconstruction quality. It introduces QABBA, which quantizes these centers to 8 bits for lengths and 12 bits for increments, cutting the storage of the representation by about half relative to ABBA while keeping runtime unchanged. The authors prove an upper bound on the extra error introduced by quantization and show empirically on standard univariate and multivariate time-series archives that QABBA's reconstruction closely tracks its non-quantized parent. As a downstream test, QABBA's symbolic chains, fed to a fine-tuned large language model, beat previous best results on 9 of 16 datasets in an extrinsic regression benchmark.

What carries the argument

The load-bearing object is the affine quantization mapping $Q(x)=\mathrm{clip}(\lfloor x/s-z\rceil,a,b)$ with scale $s=(\eta-\zeta)/(2^{\omega}-1)$ and zero-point $z$, applied componentwise to the symbolic centers. The argument works because the mean-center property makes the cross term vanish: for each cluster, $\sum_{p\in S_i}(p-c_i)^{\top}(c_i-\hat{c}_i)=0$, so the quantized SSE splits exactly into the original SSE plus $|S_i|\,\|c_i-\hat{c}_i\|^2$, bounded by $2N(\eta-\zeta)^2/(2^{\omega+1}-2)^2$. This decomposition is what lets the paper claim that quantization error is controllable and independent of the clustering algorithm used.

What would settle it

Run QABBA with the fixed 8/12-bit settings on a real-world time series whose increments span a range far wider than the synthetic Gaussian data, such as an ECG recording with occasional large spikes; if the reconstruction MSE rises well above ABBA's by more than the proved bound $2N(\eta-\zeta)^2/(2^{\omega+1}-2)^2$, or if the measured storage ratio fails to approach the predicted $\varphi_{\mathrm{QABBA}}$, the generalization claim would fail.

Watch

Extended reading notes

Core claim

QABBA is ABBA with a quantization step inserted between digitization and storage: each symbolic center, a pair (length, increment), is mapped through an affine quantization $Q(x)=\mathrm{clip}(\lfloor x/s - z\rceil, a, b)$ to an $\omega$-bit integer and inverse-quantized at reconstruction. The central claim is that this substitution is nearly free. The added error in Frobenius norm is at most $\sqrt{2k}(\eta-\zeta)/(2^{\omega+1}-2)$, and the quantized sum of squared errors is bounded by the unquantized SSE plus $2N(\eta-\zeta)^2/(2^{\omega+1}-2)^2$, so for normalized data the quantization term is negligible. With 8 bits for lengths and 12 bits for increments, QABBA halves the storage of the symbolic centers, preserves reconstruction accuracy and speed on univariate and multivariate archives, and, when its symbol chains are fed to a fine-tuned large language model, reports the best results on most datasets of the extrinsic regression benchmark.

Load-bearing premise

The choice of 8 bits for pattern lengths and 12 bits for increments, fixed after experiments on synthetic Gaussian series, is assumed to keep reconstruction error acceptable across all real-world datasets; the authors note this setting likely needs per-dataset tuning.

Editorial extensions

If this is right

  • With 8-bit lengths and 12-bit increments, the storage cost of a symbolic representation falls from $8N+(B_{\mathrm{len}}+B_{\mathrm{inc}})k+pB_T$ bits to $8N+20k+96$ bits, roughly halving the footprint for typical parameters.
  • On the standard univariate and multivariate time-series archives, QABBA's reconstruction error tracks ABBA and fABBA closely, with the small loss growing as fewer symbols are used.
  • Quantization adds no meaningful runtime overhead, so QABBA can serve as a drop-in storage optimization for existing ABBA-based pipelines.
  • Feeding QABBA's symbol chains to a fine-tuned large language model, without learning a time-series embedding, reaches the best reported results on 9 of 16 datasets in the extrinsic regression benchmark.
  • The error decomposition is generic: the same quantization-aware analysis applies to other center-based clustering or vector-quantization pipelines, a direction the paper explicitly suggests.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • One could test the representational claim directly by replacing the large language model with a much smaller linear or bag-of-symbols model; if the regression gains persist, the symbolic chain itself, not the model's scale, is doing the work.
  • Because the bit-width margins were calibrated on synthetic Gaussian data, an adaptive per-dataset bit allocator that chooses widths from the observed ranges of lengths and increments is the natural next step and would directly address the authors' caveat.
  • The paper notes that character strings dominate the remaining storage; combining QABBA with entropy coding on the symbol string could compound the savings beyond the halving reported for the centers.
  • The proved bound on quantized SSE suggests a similar per-cluster cross-term cancellation could give error guarantees for integer codebooks in neural compression or quantization-aware training, linking this result to a broader family of methods.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. The paper introduces QABBA, a quantization-based variant of ABBA for symbolic time series representation. It replaces the floating-point symbolic centers with low-bit-width integers, derives an upper bound on the induced change in the sum of squared errors (Eqs. 11–13), evaluates reconstruction accuracy and storage on UCR and UEA archives, and applies QABBA with Mistral-7B to time series regression. The main claims are improved storage efficiency with retained speed and accuracy, a provable quantization error bound, and state-of-the-art regression performance on the Monash dataset.

Significance. If the claims held, QABBA would be a useful drop-in storage optimization for ABBA with a formal error guarantee. The quantization error analysis in Section 3.3 is explicit and self-contained, and the storage analysis in Section 4 clearly identifies where the savings come from. The open-source implementation and the breadth of UCR/UEA experiments are also strengths. However, the 'retains accuracy' claim is qualified by the paper's own UEA MSE results, and the LLM regression claim rests on a dated baseline and lacks a non-quantized ABBA comparison. The core quantization idea is sound and publishable after the claims are calibrated to the evidence.

major comments (3)
  1. [§5.4 / Abstract] The abstract claims that QABBA 'retains the original speed and accuracy of symbolic reconstruction,' but Section 5.4 reports that with the fixed 8-bit/12-bit configuration chosen in Section 5.1, QABBA's approximation error 'can be significantly higher' than non-quantized ABBA/fABBA in MSE, with only DTW distance remaining similar. Since MSE is the reconstruction-error metric defined in Eq. (15), the central claim is not supported for that metric unless the bit-widths are tuned per dataset. The manuscript should either restrict the accuracy claim to DTW-based comparisons and the UCR profiles, or revise the experimental protocol to demonstrate accuracy retention on the MSE metric.
  2. [§5.1 / §3.3] The bit-width configuration is selected from 100 synthetic Gaussian series of length 5,000 and then fixed for all real-world experiments, so there is no evidence that the 8/12-bit choice generalizes to the scale, distribution, and multivariate structure of the UEA datasets. The quantization error bound in Eq. (13) does not close this gap: it bounds the change in center-space SSE, not the final reconstruction MSE, and it depends on the range (η - ζ) of the quantized centers, which is not controlled by the synthetic-data calibration. The paper should include per-dataset bit-width sensitivity or a data-driven selection rule.
  3. [§5.2 / Table 4] The claim of a 'new state-of-the-art on Monash regression dataset' is overstated. The SOTA column refers to the 2021 results of [50], not to current LLM-based regression methods, and the experiments contain no non-quantized ABBA/fABBA baseline, so the effect of quantization on regression performance is not isolated. Moreover, on several datasets (App., Hou.2, Bei.1, Flo.2, Flo.3, PPG) the QABBA results are worse than the cited SOTA, so the statement should be rephrased to 'competitive on 9 of 16 datasets' and supported by stronger baselines.
minor comments (6)
  1. [§3.3, Eqs. (9)-(11)] The notation ⌊f⌉ uses an undefined f, and Eq. (11) writes an equality where the derivation yields an upper bound; replace with |δ| ≤ (η-ζ)/(2^{ω+1}-2) and state the bound as an inequality.
  2. [§3.1, Eq. (1)] The summation 'ijX' should be typeset as \sum_{i=i_{j-1}}^{i_j}.
  3. [§3.4, Eq. (15)] The MSE formula has typographical errors; it should read MSE = (1/n) Σ_{i=1}^n (t_i - \hat t_i)^2.
  4. [§5.4] The text 'Fig. 6 and 6' should be 'Fig. 6 and Fig. 7'; also 'it likely that in practice' is missing 'is'.
  5. [§6] The sentence 'we aims to preserve' should be 'we aim to preserve'.
  6. [Table 3] The header 'Quant.' is not defined in the table body; clarify that the first column indicates whether QLoRA quantization was enabled.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the quantization error bound is self-contained, and the paper's self-citations are building blocks rather than load-bearing sources of the central claim.

full rationale

No circular steps were found. The quantization error bound in Section 3.3 is derived from the quantization map in Eqs. (6)-(8) and the rounding model in Eq. (9), independent of the paper's empirical claims; Eq. (13) follows algebraically from the mean-center identity and the componentwise rounding bound. The storage ratio in Eq. (17) is arithmetic from the representation layout. The bit-width choice in Section 5.1 is a hyperparameter setting selected on synthetic development data and then applied to UCR/UEA; it is not fitted to those test sets, so the later reconstruction evaluations are not forced by construction. Indeed, Section 5.4 candidly reports that the fixed 8/12-bit configuration can yield significantly higher MSE on UEA, which is an evidentiary weakness in the abstract's 'retains accuracy' claim rather than a circular step. Self-citations to fABBA [12], joint symbolic representation [34], and inhibition adaption [53] are used as algorithmic building blocks; none of them is the source of the quantization bound, and no uniqueness theorem is imported from the authors' prior work. The Monash SOTA statement is broader than Table 4's 9/16 result, but that is an overclaiming or correctness issue, not circularity.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The central derivation relies on the standard uniform quantization error model and on ABBA's compression and digitization properties taken from the cited literature. The bit-widths used in all experiments are free parameters selected from one synthetic dataset. No new entities are introduced.

free parameters (4)
  • len bit-width = 8 bits
    Selected based on the synthetic Gaussian experiment in Section 5.1 (safety margin [8,32)); used in all subsequent QABBA experiments.
  • inc bit-width = 12 bits
    Selected based on the synthetic Gaussian experiment in Section 5.1 (safety margin [12,32)); used in all subsequent QABBA experiments.
  • ABBA/LLM hyperparameters (scl, tol, alpha) = scl=3, tol=0.001, alpha=0.001 in LLM experiments; scl=1/5 in ECG figure
    Chosen by hand (Table 3); these affect the symbolic representation and the regression result, but are not the paper's main contribution.
  • QLoRA fine-tuning hyperparameters = 4-bit base, rank 16, alpha 16, dropout 0.3, lr 2e-4, epochs 20, batch 4
    Chosen by hand (Table 3); these are standard fine-tuning settings that the LLM regression results depend on.
assumptions (4)
  • standard math The rounding error of the quantization operator is bounded by half the step size: |delta| <= (eeta - ezeta) / (2^(omega+1) - 2), Eq. (9).
    This is the standard uniform quantization error bound and is used to derive Eqs. (11)-(13).
  • domain assumption The values of the symbolic centers lie within the quantization range [zeta, eta] after normalization, so clipping is not needed.
    The error bound omits the clip operation; if centers fall outside the range, clipping adds unmodeled error (Section 3.3).
  • domain assumption The ABBA compression criterion (Eq. 1) bounds each piece's squared error and is inherited without change.
    QABBA uses ABBA's compression and digitization; the paper does not re-derive ABBA's guarantees.
  • domain assumption The joint symbolic representation of [34] produces consistent symbols for multivariate time series.
    Section 5.4 relies on this prior work by the second author to apply QABBA to UEA multivariate datasets.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Quantized symbolic time series approximation." pith.science (2026). https://pith.science/paper/IPNYRNRP

@misc{pith2026241115209,
  author       = {Pith},
  title        = {Pith review of: Quantized symbolic time series approximation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IPNYRNRP}},
  note         = {Machine review of arXiv:2411.15209}
}
read the original abstract

Time series are ubiquitous in numerous science and engineering domains, e.g., signal processing, bioinformatics, and astronomy. Previous work has verified the efficacy of symbolic time series representation in a variety of engineering applications due to its storage efficiency and numerosity reduction. The most recent symbolic aggregate approximation technique, ABBA, has been shown to preserve essential shape information of time series and improve downstream applications, e.g., neural network inference regarding prediction and anomaly detection in time series. Motivated by the emergence of high-performance hardware which enables efficient computation for low bit-width representations, we present a new quantization-based ABBA symbolic approximation technique, QABBA, which exhibits improved storage efficiency while retaining the original speed and accuracy of symbolic reconstruction. We prove an upper bound for the error arising from quantization and discuss how the number of bits should be chosen to balance this with other errors. An application of QABBA with large language models (LLMs) for time series regression is also presented, and its utility is investigated. By representing the symbolic chain of patterns on time series, QABBA not only avoids the training of embedding from scratch, but also achieves a new state-of-the-art on Monash regression dataset. The symbolic approximation to the time series offers a more efficient way to fine-tune LLMs on the time series regression task which contains various application domains. We further present a set of extensive experiments performed across various well-established datasets to demonstrate the advantages of the QABBA method for symbolic approximation.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Signal, Image, or Symbolic: Exploring the Best Input Representation for Electrocardiogram-Language Models Through a Unified Framework

    cs.AI 2025-05 conditional novelty 5.0 of 10

    A unified benchmark across six ECG datasets and five text-generation metrics finds tokenized symbolic ECG inputs outperform raw signal and image inputs for ECG-language models.

  2. ECG-Byte: A Tokenizer for End-to-End Generative Electrocardiogram Language Modeling

    cs.CL 2024-12 conditional novelty 5.0 of 10

    A BPE-based tokenizer lets an LLM generate clinical text directly from quantized ECG signals, matching two-stage encoder methods with roughly 3x faster training and 48% of the data.

Reference graph

Works this paper leans on

57 extracted references · 26 canonical work pages · cited by 2 Pith papers

  1. [50]

    Data Mining and Knowledge Discovery 35(3), 1032–1060 (2021) https://doi.org/10.1007/ s10618-021-00745-9

    Tan, C.W., Bergmeir, C., Petitjean, F., Webb, G.I.: Time series extrinsic regression: Predicting numeric values from time series data. Data Mining and Knowledge Discovery 35(3), 1032–1060 (2021) https://doi.org/10.1007/ s10618-021-00745-9

    work page 2021

  2. [1]

    In: ACM Symposium on Theory of Computing

    Indyk, P., Motwani, R.: Approximate nearest neighbors: Towards removing the curse of dimensionality. In: ACM Symposium on Theory of Computing. STOC ’98, pp. 604–613 (1998). https://doi.org/10.1145/276698.276876

    arXiv 1998

  3. [2]

    IEEE Transactions on Knowledge and Data Engineering 29(11), 2581–2600 (2017) https://doi.org/10.1109/TKDE.2017.2740932

    Jensen, S., Pedersen, T., Thomsen, C.: Time series management systems: A sur- vey. IEEE Transactions on Knowledge and Data Engineering 29(11), 2581–2600 (2017) https://doi.org/10.1109/TKDE.2017.2740932

    arXiv 2017

  4. [3]

    ACM Computing Surveys 45(1) (2012) https://doi.org/10.1145/2379776.2379788

    Esling, P., Agon, C.: Time-series data mining. ACM Computing Surveys 45(1) (2012) https://doi.org/10.1145/2379776.2379788

    arXiv 2012

  5. [4]

    ACM Computing Surveys 55(10) (2023) https://doi.org/10.1145/3560814

    Chiarot, G., Silvestri, C.: Time series compression survey. ACM Computing Surveys 55(10) (2023) https://doi.org/10.1145/3560814

    doi:10.1145/3560814 2023

  6. [5]

    In: Proceedings of the 16th International Conference on Scientific and Statistical Database Management, pp

    Kontaki, M., Papadopoulos, A.N.: Efficient similarity search in streaming time sequences. In: Proceedings of the 16th International Conference on Scientific and Statistical Database Management, pp. 63–72. IEEE, Greece (2004). https://doi. org/10.1109/SSDM.2004.1311194

    arXiv 2004

  7. [6]

    Proceedings of the VLDB Endowment 2, 826–837 (2009) https://doi.org/10.14778/1687627.1687721

    Assent, I., Wichterich, M., Krieger, R., Kremer, H., Seidl, T.: Anticipatory DTW for efficient similarity search in time series databases. Proceedings of the VLDB Endowment 2, 826–837 (2009) https://doi.org/10.14778/1687627.1687721

    arXiv 2009

  8. [7]

    Francis, B., Dasgupta, S.: Signal compression by subband coding. Automatica 21 AtrialFibrillation BasicMotions CharacterTrajectories Epilepsy JapaneseVowels NATOPS UWaveGestureLibrary 0.0 0.1 0.2 0.3 0.4 0.5 0.6 Symbolic centers (ABBA) Strings (ABBA) Others (ABBA) Symbolic centers (QABBA) Strings (QABBA) Others (QABBA) tol = 0.005 AtrialFibrillation Basic...

    work page 1999

Show all 57 references
  1. [8]

    ACM Transactions on Database Systems 27(2), 188–228 (2002) https://doi.org/10.1145/568518.568520

    Chakrabarti, K., Keogh, E., Mehrotra, S., Pazzani, M.: Locally adaptive dimen- sionality reduction for indexing large time series databases. ACM Transactions on Database Systems 27(2), 188–228 (2002) https://doi.org/10.1145/568518.568520

    2002

  2. [9]

    In: Proceedings of the 8th ACM SIG- MOD Workshop on Research Issues in Data Mining and Knowledge Discovery, pp

    Lin, J., Keogh, E., Lonardi, S., Chiu, B.: A symbolic representation of time series, with implications for streaming algorithms. In: Proceedings of the 8th ACM SIG- MOD Workshop on Research Issues in Data Mining and Knowledge Discovery, pp. 2–11. ACM, USA (2003). https://doi.o...

    2003

  3. [10]

    In: Advances in Intelligent Data Analysis XII (2013)

    Malinowski, S., Guyet, T., Quiniou, R., Tavenard, R.: 1d-SAX: A novel sym- bolic representation for time series. In: Advances in Intelligent Data Analysis XII (2013). https://doi.org/10.1007/978-3-642-41398-8 24

    2013 doi

  4. [11]

    Data Mining and Knowledge Discovery34, 1175–1200 (2020) https://doi.org/10.1007/s10618-020-00679-8

    Elsworth, S., G¨ uttel, S.: ABBA: adaptive Brownian bridge-based symbolic aggre- gation of time series. Data Mining and Knowledge Discovery34, 1175–1200 (2020) https://doi.org/10.1007/s10618-020-00679-8

    2020 doi

  5. [12]

    ACM Transactions on Knowledge Discovery from Data (2022) https://doi.org/10.1145/3532621

    Chen, X., G¨ uttel, S.: An efficient aggregation method for the symbolic represen- tation of temporal data. ACM Transactions on Knowledge Discovery from Data (2022) https://doi.org/10.1145/3532621

    2022 doi

  6. [13]

    Data Mining and Knowledge Discovery 15(2), 107– 144 (2007) https://doi.org/10.1007/s10618-007-0064-z 22

    Lin, J., Keogh, E., Wei, L., Lonardi, S.: Experiencing SAX: a novel symbolic representation of time series. Data Mining and Knowledge Discovery 15(2), 107– 144 (2007) https://doi.org/10.1007/s10618-007-0064-z 22

    2007 doi

  7. [14]

    In: 18th International Conference on Extending Database Technology, pp

    Senin, P., Lin, J., Wang, X., Oates, T., Gandhi, S., Boedihardjo, A.P., Chen, C., Frankenstein, S.: Time series anomaly discovery with grammar-based compres- sion. In: 18th International Conference on Extending Database Technology, pp. 481–492. OpenProceedings.org, ??? (2015)....

    2015 doi

  8. [15]

    In: Advanced Analytics and Learning on Temporal Data: 7th ECML PKDD Workshop, AALTD 2022, pp

    Nguyen, T.L., Ifrim, G.: Fast time series classification with random symbolic subsequences. In: Advanced Analytics and Learning on Temporal Data: 7th ECML PKDD Workshop, AALTD 2022, pp. 50–65. Springer, France (2023). https://doi.org/10.1007/978-3-031-24378-3 4

    2023 doi

  9. [16]

    Elsworth, S., G¨ uttel, S.: Time series forecasting using LSTM networks: A symbolic approach, 12 (2020) 2003.05672

    2020 arXiv

  10. [17]

    Applied Soft Computing 122, 108871 (2022) https://doi.org/10.1016/j.asoc.2022.108871

    Criado-Ram´ on, D., Ruiz, L.G.B., Pegalajar, M.C.: Electric demand forecast- ing with neural networks and symbolic time series representations. Applied Soft Computing 122, 108871 (2022) https://doi.org/10.1016/j.asoc.2022.108871

    2022

  11. [18]

    In: Proceedings of the 10th International Workshop on Multi- media Data Mining

    Li, Y., Lin, J.: Approximate variable-length time series motif discovery using grammar inference. In: Proceedings of the 10th International Workshop on Multi- media Data Mining. MDMKDD ’10. ACM, USA (2010). https://doi.org/10.1145/ 1814245.1814255

    2010

  12. [19]

    In: 2017 IEEE International Conference on Acoustics, Speech and Signal Processing, pp

    Zhang, S., Bahrampour, S., Ramakrishnan, N., Schott, L., Shah, M.: Deep learn- ing on symbolic representations for large-scale heterogeneous time-series event prediction. In: 2017 IEEE International Conference on Acoustics, Speech and Signal Processing, pp. 5970–5974 (2017). h...

    2017 doi

  13. [20]

    In: IEEE Symposium Series on Computational Intelligence, pp

    Tamura, K., Ichimura, T.: Clustering of time series using hybrid symbolic aggre- gate approximation. In: IEEE Symposium Series on Computational Intelligence, pp. 1–8 (2017). https://doi.org/10.1109/SSCI.2017.8280846

    2017

  14. [21]

    In: IEEE International Solid-State Circuits Conference, pp

    Lee, J., Kim, C., Kang, S., Shin, D., Kim, S., Yoo, H.-J.: UNPU: A 50.6TOPS/W unified deep neural network accelerator with 1b-to-16b fully-variable weight bit- precision. In: IEEE International Solid-State Circuits Conference, pp. 218–220. IEEE, USA (2018). https://doi.org/10....

    2018

  15. [22]

    In: ACM/IEEE 43rd Annual Inter- national Symposium on Computer Architecture (ISCA), pp

    Chen, Y.-H., Emer, J., Sze, V.: Eyeriss: A spatial architecture for energy-efficient dataflow for convolutional neural networks. In: ACM/IEEE 43rd Annual Inter- national Symposium on Computer Architecture (ISCA), pp. 367–379 (2016). https://doi.org/10.1109/ISCA.2016.40

    2016 doi

  16. [23]

    In: Proceedings of the 50th Annual IEEE/ACM International Symposium on Microarchitecture

    Albericio, J., Delm´ as, A., Judd, P., Sharify, S., O’Leary, G., Genov, R., Moshovos, A.: Bit-Pragmatic deep neural network computing. In: Proceedings of the 50th Annual IEEE/ACM International Symposium on Microarchitecture. MICRO-50 ’17, pp. 382–394. ACM, USA (2017). https://...

    2017

  17. [24]

    ACM Transactions on Intelligent Systems and Technology 14(6) (2023) https://doi.org/10.1145/ 3623402

    Rokh, B., Azarpeyvand, A., Khanteymoori, A.: A comprehensive survey on model quantization for deep neural networks in image classification. ACM Transactions on Intelligent Systems and Technology 14(6) (2023) https://doi.org/10.1145/ 3623402

    2023

  18. [25]

    In: International Conference on Automation, Computing and Renew- able Systems, pp

    Kulkarni, U., Hosamani, A.S., Masur, A.S., Hegde, S., Vernekar, G.R., Siri Chan- dana, K.: A survey on quantization methods for optimization of deep neural networks. In: International Conference on Automation, Computing and Renew- able Systems, pp. 827–834 (2022). https://doi....

    2022

  19. [26]

    In: 2019 IEEE Visual- ization Conference, pp

    Ruta, N., Sawada, N., McKeough, K., Behrisch, M., Beyer, J.: SAX navigator: Time series exploration through hierarchical clustering. In: 2019 IEEE Visual- ization Conference, pp. 236–240. IEEE, Canada (2019). https://doi.org/10.1109/ VISUAL.2019.8933618

    2019

  20. [27]

    Computers & Graphics 112, 13–21 (2023) https://doi.org/10.1016/j

    Yu, Y., Becker, T., Trinh, L.M., Behrisch, M.: SAXRegEx: Multivariate time series pattern search with symbolic representation, regular expression, and query expansion. Computers & Graphics 112, 13–21 (2023) https://doi.org/10.1016/j. cag.2023.03.002

    2023 doi

  21. [28]

    Data Mining and Knowledge Discovery 35(6), 2369–2388 (2021) https://doi.org/10

    Li, X., Lin, J., Zhao, L.: Time series clustering in linear time complexity. Data Mining and Knowledge Discovery 35(6), 2369–2388 (2021) https://doi.org/10. 1007/s10618-021-00798-w

    2021

  22. [29]

    In: IEEE International Conference on Data Mining

    Keogh, E., Lin, J., Fu, A.: HOT SAX: efficiently finding the most unusual time series subsequence. In: IEEE International Conference on Data Mining. ICDM’05, pp. 1–8 (2005). https://doi.org/10.1109/ICDM.2005.79

    2005 doi

  23. [30]

    In: Proceedings of the 8th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

    Keogh, E., Lonardi, S., Chiu, B.Y.-c.: Finding surprising patterns in a time series database in linear time and space. In: Proceedings of the 8th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. KDD ’02, pp. 550–556. ACM, USA (2002). https://doi.org/...

    2002

  24. [31]

    In: IEEE International Conference on Data Mining, pp

    Senin, P., Malinchik, S.: SAX-VSM: Interpretable time series classification using SAX and vector space model. In: IEEE International Conference on Data Mining, pp. 1175–1180 (2013). https://doi.org/10.1109/ICDM.2013.62

    2013 doi

  25. [32]

    In: IEEE International Conference on Data Mining, pp

    Li, X., Lin, J.: Linear time complexity time series classification with bag-of- pattern-features. In: IEEE International Conference on Data Mining, pp. 277–286 (2017). https://doi.org/10.1109/ICDM.2017.37

    2017 doi

  26. [33]

    IEEE Transactions on Knowledge and Data Engineering 35(6), 5752–5766 (2023) https://doi.org/10.1109/TKDE.2022.3141158 24

    Bountrogiannis, K., Tzagkarakis, G., Tsakalides, P.: Distribution agnostic sym- bolic representations for time series dimensionality reduction and online anomaly detection. IEEE Transactions on Knowledge and Data Engineering 35(6), 5752–5766 (2023) https://doi.org/10.1109/TKDE...

    2023

  27. [34]

    Chen, X.: Joint symbolic aggregate approximation of time series (2024) 2401.00109

    2024 arXiv

  28. [35]

    IEEE ASSP Magazine 1(2), 4–29 (1984) https: //doi.org/10.1109/MASSP.1984.1162229

    Gray, R.: Vector quantization. IEEE ASSP Magazine 1(2), 4–29 (1984) https: //doi.org/10.1109/MASSP.1984.1162229

    1984

  29. [36]

    IEEE Transactions on Information Theory 55(7), 3229–3242 (2009) https://doi.org/10

    Dasgupta, S., Freund, Y.: Random projection trees for vector quantization. IEEE Transactions on Information Theory 55(7), 3229–3242 (2009) https://doi.org/10. 1109/TIT.2009.2021325

    2009

  30. [37]

    IEEE Transactions on Information Theory 28(2), 129–137 (1982) https://doi.org/10.1109/TIT.1982.1056489

    Lloyd, S.: Least squares quantization in PCM. IEEE Transactions on Information Theory 28(2), 129–137 (1982) https://doi.org/10.1109/TIT.1982.1056489

    1982

  31. [38]

    Machine Learning 56(1-3), 9–33 (2004) https://doi.org/10.1023/B:MACH.0000033113.59016.96

    Drineas, P., Frieze, A., Kannan, R., Vempala, S., Vinay, V.: Clustering large graphs via the singular value decomposition. Machine Learning 56(1-3), 9–33 (2004) https://doi.org/10.1023/B:MACH.0000033113.59016.96

    2004

  32. [39]

    Theoretical Computer Science 442, 13–21 (2012) https://doi.org/ 10.1016/j.tcs.2011.12.034

    Mahajan, M., Nimbhorkar, P., Varadarajan, K.: The planar k-means problem is np-hard. Theoretical Computer Science 442, 13–21 (2012) https://doi.org/ 10.1016/j.tcs.2011.12.034 . Special Issue on the Workshop on Algorithms and Computation (W ALCOM 2009)

    2012 doi

  33. [40]

    In: Proceedings of the Ninth IEEE International Conference on Computer Vision, vol

    Yu, S.X., Shi, J.: Multiclass spectral clustering. In: Proceedings of the Ninth IEEE International Conference on Computer Vision, vol. 2, p. 313. IEEE, ??? (2003). https://doi.org/10.1109/ICCV.2003.1238361

    2003 arXiv

  34. [41]

    In: Proceedings of the Second International Conference on Knowledge Discovery and Data Mining

    Ester, M., Kriegel, H.-P., Sander, J., Xu, X.: A density-based algorithm for discov- ering clusters in large spatial databases with noise. In: Proceedings of the Second International Conference on Knowledge Discovery and Data Mining. KDD’96, pp. 226–231. AAAI Press, USA (1996)

    1996

  35. [42]

    In: Advances in Knowledge Discovery and Data Mining, pp

    Campello, R.J.G.B., Moulavi, D., Sander, J.: Density-based clustering based on hierarchical density estimates. In: Advances in Knowledge Discovery and Data Mining, pp. 160–172. Springer, Germany (2013). https://doi.org/10.1007/ 978-3-642-37456-2 14

    2013

  36. [43]

    org/10.1016/j.patcog.2019.04.014

    Fr¨ anti, P., Sieranoja, S.: How much can k-means be improved by using better initialization and repeats? Pattern Recognition 93, 95–112 (2019) https://doi. org/10.1016/j.patcog.2019.04.014

    2019 doi

  37. [44]

    Pattern Recognition 110, 107625 (2021) https://doi.org/10.1016/j.patcog.2020.107625

    Saha, J., Mukherjee, J.: CNAK: cluster number assisted k-means. Pattern Recognition 110, 107625 (2021) https://doi.org/10.1016/j.patcog.2020.107625

    2021

  38. [45]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2018)

    Jacob, B., Kligys, S., Chen, B., Zhu, M., Tang, M., Howard, A., Adam, H., Kalenichenko, D.: Quantization and training of neural networks for effi- cient integer-arithmetic-only inference. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2018)....

    2018

  39. [46]

    IEEE Transactions on Acoustics, Speech, and Signal Processing 26(1), 43–49 (1978) https://doi.org/10.1109/TASSP.1978.1163055

    Sakoe, H., Chiba, S.: Dynamic programming algorithm optimization for spoken word recognition. IEEE Transactions on Acoustics, Speech, and Signal Processing 26(1), 43–49 (1978) https://doi.org/10.1109/TASSP.1978.1163055

    1978

  40. [47]

    IEEE/CAA Journal of Automatica Sinica 6(6), 1293–1305 (2019) https://doi.org/10.1109/ JAS.2019.1911747

    Dau, H.A., Bagnall, A., Kamgar, K., Yeh, C.-C.M., Zhu, Y., Gharghabi, S., Ratanamahatana, C.A., Keogh, E.: The UCR time series archive. IEEE/CAA Journal of Automatica Sinica 6(6), 1293–1305 (2019) https://doi.org/10.1109/ JAS.2019.1911747

    2019

  41. [48]

    CoRR (2018)

    Bagnall, A.J., Dau, H.A., Lines, J., Flynn, M., Large, J., Bostrom, A., Southam, P., Keogh, E.: The UEA multivariate time series classification archive. CoRR (2018)

    2018

  42. [49]

    In: SODA ’07: Proceedings of the 8th Annual ACM-SIAM Symposium on Discrete Algorithms, pp

    Arthur, D., Vassilvitskii, S.: k-means++: the advantages of careful seeding. In: SODA ’07: Proceedings of the 8th Annual ACM-SIAM Symposium on Discrete Algorithms, pp. 1027–1035. Society for Industrial and Applied Mathematics, ??? (2007)

    2007

  43. [51]

    arXiv preprint arXiv:2310.06825 (2023)

    Jiang, A.Q., Sablayrolles, A., Mensch, A., Bamford, C., Chaplot, D.S., Casas, D.d.l., Bressand, F., Lengyel, G., Lample, G., Saulnier, L., et al.: Mistral 7b. arXiv preprint arXiv:2310.06825 (2023)

    2023 arXiv

  44. [52]

    Advances in Neural Information Processing Systems 36 (2024)

    Dettmers, T., Pagnoni, A., Holtzman, A., Zettlemoyer, L.: QLoRA: Efficient fine- tuning of quantized LLMs. Advances in Neural Information Processing Systems 36 (2024)

    2024

  45. [53]

    Neural Networks, 106410 (2024) https: //doi.org/10.1016/j.neunet.2024.106410

    Kang, C., Prokop, J., Tong, L., Zhou, H., Hu, Y., Novak, D.: InA: Inhibition adaption on pre-trained language models. Neural Networks, 106410 (2024) https: //doi.org/10.1016/j.neunet.2024.106410

    2024

  46. [54]

    Advances in Neural Information Processing Systems 32 (2019)

    Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., et al.: PyTorch: An imperative style, high- performance deep learning library. Advances in Neural Information Processing Systems 32 (2019)

    2019

  47. [55]

    In: International Conference on Learning Representations, Austria (2024) 26

    Jin, M., Wang, S., Ma, L., Chu, Z., Zhang, J.Y., Shi, X., Chen, P.-Y., Liang, Y., Li, Y.-F., Pan, S., Wen, Q.: Time-LLM: Time series forecasting by reprogramming large language models. In: International Conference on Learning Representations, Austria (2024) 26

    2024

  48. [56]

    Mathematical Programming 91(2), 201–213 (2002) https://doi.org/10

    Dolan, E.D., Mor´ e, J.J.: Benchmarking optimization software with performance profiles. Mathematical Programming 91(2), 201–213 (2002) https://doi.org/10. 1007/s101070100263

    2002

  49. [57]

    Computer 17(6), 8–19 (1984) https://doi.org/10.1109/MC.1984.1659158 27

    Welch, T.A.: A technique for high-performance data compression. Computer 17(6), 8–19 (1984) https://doi.org/10.1109/MC.1984.1659158 27

    1984

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.